Dynamic locomotion of quadrupedal robots over rough terrain

Abstract: Previous works have enabled locomotion of quadrupedal robots using the ZMP-based motion optimization framework on flat terrain with various gait patterns. In this work, terrain perception is integrated into the ZMP-based motion optimization framework to enable robots to perform dynamic locomotion over rough terrain. In a first step, we extend the foothold optimization framework to use processed terrain information to avoid planning unsafe foothold positions while traversing over rugged terrain. In order to avoid contact slippage, we drop the approximated terrain normal and use measured terrain normal at foot contact position to compute the friction cone. Further, to avoid kinematic violations during locomotion over rugged terrain, we present modifications to the ZMP-based motion optimization framework to solve for kinematically feasible motion plans in realtime. We add nonlinear kinematic constraints to existing nonlinear ZMP motion optimization framework and solve a Sequential Quadratic Programming (SQP) problem to obtain feasible motion plans. The proposed algorithms is tested in simulation and on hardware with various gaits to show that the robot can traverse rugged terrain safely. Finally, the computational time and performance of the proposed algorithms were analyzed under various scenarios and presented as part of this work.

Examiner: Prof. Elling W. Jacobsen